Data Engineer & ML Engineer — I build the pipelines that make machine learning actually work in production.

4+ years of experience across enterprise data engineering (Accenture) and ML research (UNO). I specialize in designing reliable ETL workflows, end-to-end ML pipelines, and data systems that hold up under real-world conditions.

Languages & Data Python · SQL · PL/SQL · Pandas · NumPy

ML & AI PyTorch · scikit-learn · TensorFlow · OpenCV · PyWavelets

Data Engineering ETL Pipelines · Batch Processing · Data Warehousing · Schema Design · Data Quality

Cloud & Tools AWS (S3, EC2, Glue) · Azure (Data Factory, Databricks) · Snowflake · Docker · Git · Linux

BI & Visualization Power BI · Tableau

Fusing CT and MRI brain scans using Dual-Tree Complex Wavelet Transform — end-to-end Flask web app with landmark-based registration, wavelet fusion, and watershed segmentation.

• Evaluated across 20+ scan pairs using PSNR, entropy, and fusion factor metrics
• Published: IJEAST 2022 — DOI: 10.33564/IJEAST.2022.v06i12.054

Python Flask OpenCV NumPy PyWavelets

Predicting network throughput from RTT and jitter using 5 regression models on 600 samples generated from Mininet emulation on Azure — Random Forest achieved MAE of 6.18 Mbps, Decision Tree R² of 0.497.

• Full pipeline: data collection → EDA → training → evaluation → diagnostic visualizations

Python scikit-learn Mininet Azure Pandas Matplotlib

End-to-end Retrieval-Augmented Generation pipeline that answers natural language questions about my capstone research — PDF ingestion, semantic chunking, FAISS vector search, and a Hugging Face QA model deployed as a live Streamlit web app.

• 94 chunks embedded using Sentence Transformers (all-MiniLM-L6-v2)
• Semantic retrieval via FAISS with 384-dimensional vectors
• Live demo: raja-rag-research.streamlit.app

LangChain FAISS Sentence Transformers Hugging Face Streamlit Python

Data & ML Research Engineer — University of Nebraska at Omaha (2024 – Present)

• Built end-to-end IoT data pipeline processing 10,000+ sensor records daily across 43 attributes
• Achieved AUC-ROC of 0.999 on adversarial robustness detection using PyTorch autoencoders
• Reduced data quality failures by ~69.6% via batch-level validation and input sanitization

Data Engineer — Accenture, Client: Best Buy (2022 – 2024)

• Designed Oracle EBS batch workflows processing 500K+ daily financial transactions
• Improved query throughput by 15–20% via execution plan analysis and strategic indexing
• Saved 5+ hours/sprint through Python automation for data validation and reconciliation

• M.S. Computer Science — University of Nebraska at Omaha (GPA: 3.67, May 2026)
• Publication: K. Rajasekhar et al., "Medical Image Fusion using Dual-Tree Complex Wavelet Transform for CT and MRI Modalities" — IJEAST 2022

LinkedIn · Email

Actively looking for Data Engineer and ML/AI Engineer roles — open to remote and hybrid.